Skip to content

Correct the radio matrix in the README - #140

Merged
marknolan merged 1 commit into
mainfrom
fix/readme-radio-matrix
Sep 14, 2026
Merged

marknolan merged 1 commit into
mainfrom
fix/readme-radio-matrix

Conversation

@marknolan

Copy link
Copy Markdown
Member

Follow-up to the review comment on #139, which flagged that the RN42/RN4678 description
over-simplifies which revisions carry which radio. The same text is here in the README — in fact
this is where the CLAUDE.md wording came from — so it needs fixing at source.

Two corrections:

1. The revision cutoff. The README said boards up to expansion-board revision 5 carry an RN42
and revision 6+ an RN4678. That isn't how it works: the transition tracks sensor generation, the
revision gates differ per board ID (SR31 / SR38 / SR47 / SR48 / SR49 each run their own
major/minor scheme), and assembly variants of one PCB are encoded as a higher minor revision — so
a bare revision number means nothing without its board ID. It now points at
SHIMMER3_BOARD_REVISIONS.md
rather than restating a matrix this file can't keep current. The runtime detail worth keeping (the
two EEPROM bits in ShimBt_startCommon, and no-EEPROM meaning the RN42 fleet) stays.

2. The Shimmer3R radio. The device table listed Shimmer3R as nRF52 (BLE) + RN4678. It is
neither — it carries a CYW20820 (Vela IF820). This repo's own source already says so
(WiredShimmerClient.ts parses "the CYW20820's version record", and brandRecord.ts distinguishes
the RN4678's 31-byte advertisement from Shimmer3R's full payload); shimmer3r-firmware and the
board-revisions doc agree. The "native nRF52 BLE" comparison further down is corrected to match.

Worth a check from someone with the PCB index workbook: the SR31-6-0 onward gate in the table is
left as-is, since I had no way to verify it and it is at least board-ID-specific.

🤖 Generated with Claude Code

Two errors, both of which the CLAUDE.md PR inherited by paraphrasing them.

The RN42/RN4678 split was written as a single "expansion-board revision 5 vs
6+" cutoff. It isn't: the transition tracks sensor generation, the gates differ
per board ID, and assembly variants of one PCB are encoded as higher minor
revisions — so a bare revision number means nothing without its board ID. Now
points at SHIMMER3_BOARD_REVISIONS.md instead of restating a matrix this file
cannot keep current.

The device table also listed Shimmer3R as "nRF52 (BLE) + RN4678". It is neither
— it carries a CYW20820 (Vela IF820). This repo's own source already says so:
WiredShimmerClient parses "the CYW20820's version record", and the firmware and
board-revisions doc agree. The "native nRF52 BLE" comparison further down is
corrected to match.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

@claude claude Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Claude Code Review

This repository is configured for manual code reviews. Comment @claude review for a one-time review, or @claude review always to subscribe this PR to a review on every future push.

Tip: disable this comment in your organization's Code Review settings.

@marknolan
marknolan merged commit 5d04839 into main Sep 14, 2026
2 checks passed
@marknolan
marknolan deleted the fix/readme-radio-matrix branch September 14, 2026 11:22
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant